Project Summary The term autism-spectrum disorders (ASD) exemplifies the tremendous heterogeneity in this developmental disorder at both the phenotypic and underlying genetic levels. It has repeatedly been observed that ASD disproportionately affects males (B) relative to females (@). Although many hypotheses attempt to explain this bias, no clear answers have emerged because of inconsistent and incomplete phenotyping and small sample sizes. We propose to leverage the interdisciplinary strengths and recruiting power of our network to study sex- specific differences by deep phenotyping and genotyping of ASD participants. We will recruit a sex-balanced cohort of ASD (N=125 B N=125 @) and matched typically developing (TD) comparison participants (N=125 B, N=125 @), as well as a set of unaffected siblings (US; N=63 @, N=62 B). We will quantitatively phenotype multiple behavioral domains and measure several key ASD-related neural systems at the level of brain structure (sMRI), connectivity (DTI and fMRI), function (task based and resting state fMRI), and temporal dynamics (EEG). Additionally, we will measure copy number variation (CNV) and single nucleotide variation (SNV) for these participants and their parents, allowing us to test sex- and circuit-specific genotype-phenotype hypotheses for five candidate ASD genes and ultimately extend our methods to a search for novel sex-specific and high-risk genes. Our Specific Aims are to: 1) Identify sex differences in brain structure, function, connectivity, and temporal dynamics in ASD. 2) Characterize associations between DNA sequence and copy number variants and brain structure and function in @ASD and @TD versus BASD and BTD. 3) Relate brain differences in structure, function, and temporal dynamics to heterogeneity in ASD behavior and genetics. We hypothesize that advanced network methods can aid in understanding the tremendous heterogeneity in ASD by connecting different levels of phenotype with genetic variation. We will therefore combine multiple levels of biology and endophenotypes - SNVs, CNVs, behavioral metrics, and resting state imaging and electrophysiology measures - into one framework across affected and unaffected siblings and controls using an integrated network analysis, iWGCNA.